PROJECT SUMMARY In this project we will bridge the traditionally largely distinct fields of quantitative genetics and mechanistic biology to obtain a mechanistic understanding of regulatory effects of genetic variants in humans. Leveraging on large human data sets providing parallel whole genome and transcriptome sequencing data, we will extend proof-of-principle studies and computational approaches developed and validated in model organisms to achieve improved functional interpretation of GWAS loci associated to mental health disorders. We focus specifically on the role of transcription factors as both upstream regulators of genetic risk variants as well as mediators of downstream network-level effects. As Aim 1, we will develop extend methods to allow accurate modeling of transcription factor activity from transcriptome data from large cohorts of human tissue samples in GTEx, PsychENCODE, and TOPMed cohorts. These data will be used in Aim 2 to dissect the mechanisms underlying proximal genetic regulatory variants in cis. We hypothesize that dynamics of transcription factor activity and binding modifies the effect size of genetic regulatory variants across individuals, tissues, and cell types, and that by modeling this relationship we can detect TFs regulating specific regulatory variants and noncoding disease-associated loci. In parallel Aim 3, we will map network-level trans-acting genetic variants for inter-individual variation in TF activity. Going beyond treating TF activity as a tissue-specific parameter of the cellular environment, we will now consider it as a variable quantitative trait itself, and by GWAS/TWAS for inferred TF activity, we map specific polymorphisms that affect TF activity within each tissue. We anticipate that the trans-acting loci discovered in this analysis will be of major interest not only to basic biology of regulatory networks, but also for explaining GWAS associations to complex diseases, and to mental health in particular.